With over a third of small businesses adopting AI*, others feel pressure to do the same. They worry that if they don’t use it, they risk being left behind. But adoption isn’t simple, as nearly half of small businesses agree AI is moving too fast to know when to invest*.
About 1 in 3
small businesses are adopting AI.
Almost half
of small businesses agree AI is moving too fast to know when to invest.
Source: National AI Centre AI Adoption Tracker, businesses with 0–19 employees, accessed October 2026
The challenge, then, is finding out what AI can do for your business and where to start.
We sat down with Sam Allert, CEO of Reckon, who believes the risk is real: “If you’re not embracing AI in business, you’re at a massive disadvantage.” We discussed whether businesses really need AI, where to start, what to automate, and what to keep human.
This interview was edited for clarity and length.
Does my business need AI?
The only way to know whether AI is right for your business is to explore it. AI is a powerful tool that businesses can’t ignore, Allert says. But the decision is up to them, “If they think there’s any area of their business or their life that they can automate or optimise, then they should review AI.”
AI is just a tool. It can’t replace the “Intuition, gut instinct, that human connection” needed to run your business.
Allert used a gardening business as an example: “It’s not going to do the mowing and the whipper snipping and all of that,” but AI could be “automating admin, it might be automating marketing, it might be automating invoices out.”
Explore it first, then decide whether AI is worth investing in.
Where should I start with AI?
The best place to start is training. About one in four small businesses say they don’t know how to use AI*. As Allert suggests, businesses shouldn’t “… try to solve it on their own. Invest in a one-hour course. It could be online, or a day workshop.”
About 1 in 4
small businesses say they’re not aware of how to use AI in their business.
Source: National AI Centre AI Adoption Tracker, businesses with 0–19 employees, accessed October 2026
This is what Allert did with Reckon. He took the management team on a one-day AI workshop, and it “just opened our eyes to possibilities” for what it could do for the business. Each team and department then did role-specific training. Product development saw the biggest payoff: a 30% efficiency gain with the same team.
The value in training is learning in a group setting, as Allert says:
“There’ll be people in that room, probably in the same space as you, and someone over lunch will say, ‘wow, I never thought I could automate my marketing lead capture with AI’”
Sharing ideas with like-minded people and solving problems together is how you can find out where to start with AI. Go book a course, learn what others are doing, and see how AI fits your business.
What should I automate first?
Start with your biggest bottleneck. Allert believes there is no one-size-fits-all approach to AI, and businesses shouldn’t automate the same things. Every business has its own problems: “What’s a bottleneck in your business? What’s a bottleneck in your life?” If parts of your business work well, don’t start there.
Allert used a tradie on the tools all day as an example: “If you get to the end of the day and have to read 50 emails because you’re worried about missing one, look at that for automating.” Now compound that with other admin tasks. Allert sees a major drain for small businesses in repetitive, time-consuming tasks like matching receipts and chasing invoices.
So where should businesses automate first? Automate the task that causes the biggest headache. After a month, review whether automating that bottleneck saved you time.
What should I keep doing myself?
Business owners need to keep what they’re accountable for. In Allert’s role, he receives a lot of emails and messages, which he knows AI can handle: “I know that AI could automate that and give me a summary. I don’t do that.” If something important slips through, “it’s my responsibility not to miss it.” AI may miss or overlook what you find important.
Among small businesses using or planning to use AI, only about one in four checks or plans to check AI results before they affect customers*. Checking AI output is crucial to staying accountable.
Only about 1 in 4
small businesses using or planning to use AI check, or plan to check, AI results before they affect customers.
Source: National AI Centre AI Adoption Tracker, businesses with 0–19 employees, accessed October 2026
Reckon’s AI support chat has handled over 10,000 conversations, but 75.6% of customer contact is phone calls, against 7.4% for AI chat. Most of Reckon’s customers still choose to talk to a person, something AI can’t replace.
Keeping yourself accountable requires you to ask three questions:
- If this goes wrong, does it land with me?
- Does this task need my judgement?
- Are my customers or clients expecting to deal with me?
If the answer to any of these is yes, these tasks need to stay with you and be checked before they go out.
Is there a risk to using AI?
AI poses a real risk that many aren’t seeing. You stop thinking for yourself. Allert pictures your comfort zone as a circle. You keep challenging yourself, and your comfort zone, your skill base grows. If you stop challenging yourself, your skills and comfort zone will shrink. Hand everything to AI, and “you won’t notice for one or two years but will notice in five or 10 years.”
He saw this recently with Reckon’s budget. “Simply put, the budget just wasn’t stacking up for next year.” Allert spent most of the week working through it and, within two days, had four or five options to explore.
AI could have produced a budget for him, but “I would have lacked all of that critical thinking and strategic thinking.” Working through it himself did more than produce options: “if anything, it’s energised me.”
Outsourcing your critical thinking is a mistake. Hand AI the repetitive work and keep the thinking that steers your business.
Will small businesses be left behind?
Not if they explore AI and decide for themselves whether it helps. This might mean deploying AI across different areas of your business, or in just one; “maybe it doesn’t change their business at all,” but it comes down to choice. If AI doesn’t deliver meaningful outcomes, they can “carry on”, as Allert puts it. Ultimately, AI is just a tool.
As Allert says, “AI won’t make the job easier. It’ll hand back time. What you do with that time is the impact.”
Impact over output: the hours AI frees up matter only if you spend them with purpose, for your customers, and for yourself.
























